When an organisation needs a paid Diagnose before building a system
Five signs that the quote you received is still a guess, and what a discovery engagement should actually produce.
Operate
Tim MarkasDevTechnology Partner, Jakarta25 September 2026 · 2 min read
Maintenance vs Operate: why "fix it when it breaks" is not enough for critical systems
What a retainer for a critical system should include: an SLA, change windows, backup verification, and reports leadership actually reads.
Tim MarkasDevTechnology Partner, Jakarta
A maintenance contract often says something simple: the vendor will fix errors when they happen. For a supporting application, that is enough. For a system that stops operations when it goes down, the model leaves three problems.
First, nobody watches before users complain. Second, small changes pile up because there is no clear schedule. Third, leadership never sees the state of the system until a major incident.
What makes Operate different
Operate is a retainer with written responsibilities. At MarkasDev the default scope is:
- basic monitoring and alert triage for registered systems,
- scheduled patching and maintenance for the components we manage,
- backup verification and periodic restore tests,
- incident coordination,
- small changes within an allowance, and
- monthly reports and a quarterly roadmap.
The limits are written down too. Major rebuilds, third-party licence procurement, and incidents caused by changes outside change control are excluded unless agreed separately.
A monthly rhythm
A good retainer has a predictable rhythm:
| Week | Activity |
|---|---|
| 1 | Health check and ticket backlog review |
| 2 | Maintenance window and patching |
| 3 | Rotating review of risk, vendors, or cost |
| 4 | Report to the sponsor with proposed priorities |
This rhythm makes change happen on a plan, and risks show up before they become incidents.
A sensible SLA
The SLA is agreed per contract by severity. For example: P1 for production down or corrupted critical data, P2 for an impaired critical feature, P3 for non-critical disruption, and P4 for small requests that go to the backlog. Response times are agreed together, not copied from a template.
A checklist for your contract
Before signing a retainer, make sure the contract answers:
- Which systems and components are on the asset list?
- What is the escalation path, and who is called for a P1?
- How often are backups restore-tested?
- How large is the monthly allowance for small changes?
- Which reports does leadership receive, and when?
If your system only has a reactive maintenance contract today, request Operate retainer options. We start by checking the state of the system before drafting the asset list and SLA.
Related questions
Does Operate guarantee zero downtime?
No. Operate makes sure incidents are detected, coordinated, and reported according to the agreed SLA.
When should Operate start?
Ideally at go-live, with the framework prepared in the Deliver proposal.
